Service · 35 schedules in production
AI consulting and automation your team runs on its own afterwards
AI in daily work: skills, tools and workflows your team runs on its own afterwards.
My Claude skill for German-language SEO copy passes 34 of 34 checks; without it, 23. My server runs 35 schedules, two of them with Claude: a weekly security audit and a bug hunt through the entire code base, both read-only. I set up AI for your team the same way: as a skill, as a tool on your own data, or as a workflow that runs overnight and reports in the morning.
A week on my server: 35 schedules, two of them with AI
All 35 schedules
| Application | Schedule (UTC) | Task |
|---|---|---|
| SEO portal | Mon 00:30 | Weekly export and weekly report |
| daily 02:30 | Daily Search Console and GA4 sync | |
| daily 03:50 | Technical data: URL inspection and sitemaps | |
| Sun 03:30 | Core Web Vitals and PageSpeed | |
| daily 05:10 | SEO database backup with integrity check | |
| every 10 minutes | Onboard new properties, backfill history | |
| Operations & security | hourly | Back up and compare SSH keys |
| hourly | Mirror the report drop into the archive | |
| on every reboot | Checklist after every reboot | |
| weekly | Vulnerability check of all Node projects | |
| weekly | Weekly check: processes, attacks, disk, updates, backups | |
| twice a week | Check system file integrity | |
| twice a week | Check SSH configuration for lock-out risk | |
| twice a week | Report uncommitted changes in all repos | |
| twice a week | Report updates that need a reboot | |
| twice a week | Disk usage against an 80 % threshold | |
| twice a week | Alert digest: send findings in one batch | |
| daily | Detect hung or silent jobs | |
| weekly | AI security audit of the running server AI | |
| weekly | Lint the crontab for duplicates | |
| weekly | AI bug hunt through the source code, file by file AI | |
| weekly | Update Cloudflare IP ranges | |
| on the 1st of the month | External port scan against my own server | |
| on the 1st of the month | Find orphaned Cloudflare Workers | |
| on the 1st of the month | Monthly maintenance with a backup restore test | |
| SOVP | daily 02:00 | Full backup of all databases to external storage |
| daily 03:00 | Delete PDF reports after 30 days | |
| daily 03:00 | Privacy clean-up run | |
| twice a week | Five security regression tests against the engine | |
| daily 05:00 | Monitor customer domains for score drops | |
| CERTavia | daily 02:00 | Daily copy of the customer database |
| daily 03:00 | Clean up the evidence vault by retention period | |
| daily 04:00 | Monitor customer domains for score drops | |
| Tue, Thu 07:00 | Find stuck onboardings after purchase | |
| QuickScan | daily 03:15 | Delete lead addresses after seven days |
What it covers
- Claude skills and workflows for your teamSkill test: 34 of 34 with the skill, 23 without ↗Verifiable
- Your own AI tools on your data: text from the AI, numbers from measured sourcesShown live in a call →Demo
- Advice on AI workflowsExamples in a call →Demo
- AI as an auditor in operations: weekly security audit of the serverShown live in a call →Demo
- AI bug hunt through the entire source code, file by fileShown live in a call →Demo
- Guardrails for AI: read-only access, no automatic model calls in customer applicationsShown live in a call →Demo
- Claude API: analysis, drafts, web search with sourcesShown live in a call →Demo
- Local AI tools without cloud syncShown live in a call →Demo
- Scheduled pipelines and remindersShown live in a call →Demo
- Domain monitoring with alerts on score dropsShown live in a call →Demo
- Retention periods as schedules: lead data after seven days, daily clean-up runsShown live in a call →Demo
- Counter-check and learning from outcomesShown live in a call →Demo
- macOS automation (Mail, Calendar, Contacts)Shown live in a call →Demo
Verifiable: public source. Measurable: check it yourself with an independent tool. Demo: runs in my own systems and is shown in a call.
Built with
- Claude API
- cron
- systemd
- AppleScript
Connected to
AI Consulting & Automation
- Technical SEO
- SEO ReportingReports run on their own
- GEO & AI Visibility
- Agentic SEO
- Custom SoftwareSoftware with agents
- Web Development & Operations
Frequently asked questions
How much does AI consulting cost?
The entry point is a workshop with your team: €1,200 for a half day, €2,000 for a full day. All prices net. All prices →
What stays with us after the consulting?
Not slides but tools that run: skills, workflows and schedules, documented and set up so your team operates and adapts them itself.
Which AI do you use, and what happens to our data?
By default Anthropic's Claude via the API. Anthropic does not use API data for training. If your requirements call for it, I also work with OpenAI or Gemini, and with my tools you choose which interface to connect. Where read access is enough, I set up workflows with read access only, and we sign a data processing agreement first.
Is AI automation worth it for small teams?
Yes. My own business is a one-person company with 35 schedules on the server that work overnight and report in the morning. Small teams in particular gain time that would otherwise go into routine.
How we work together
A first call costs nothing. Afterwards you know whether and how we work together.